I was listening to podcast 411 while I walked into work this morning. It was all about advertising on podcasts. In particular there is a company called Kiptronic that will dynamically insert ads into your podcast when it is being downloaded. It sounds like the service is in its infancy but I think it has huge potential. It has the potential to allow any podcaster to easily generate advertising revenue from their podcast and also to allow advertisers to easily advertise on a large number of podcasts.

At the moment the advertiser can target the ads based on geography. I presume this is done using the IP address of the person downloading the podcast. If other demographic information can be collected then the ads could be even more targeted. This is not unlike Google Adwords and Adsense - where I am shown ads that are not only relevant to my search term or content I’m reading but also to my location.

I saw Google recently got into the radio ad business with the purchase of dMarc Broadcasting for somewhere between $102 million and $1.1 billion. I’m not sure whether this technology could be applied to podcasts or not - but I’m sure Google are keeping an eye on what Kiptronics are dong.

Audio ads are more difficult to track than web based ads so it will be difficult to get the same performance measures as you have for web based advertising but I can see the future where advertisers are bidding on audio ads for podcasts/radio in much the same was as the do for web ads now. The result will be more relevant ads for the listener and better returns for the advertisers.

Paid search is simple, right? You just chose your keywords, write your ad, decide how much you want to spend and voila - you’re getting traffic… and paying money. Whereas search engine optimization is hard. The search engines can’t let you know exactly what you’ve got to do to rank well so you have to do a lot of research to find out what works and what doesn’t work. It’s different for every search engine, it’s changing and there’s a lot of information out there that’s wrong. If you don’t want to do it yourself you can hire an agency.

The Search Engine Marketing Professionals Organization (of which SLI is a member) recently released a survey on search engine marketing spending that found advertisers in US and Canada spent $5.75 billion in 2005. 83% of that was on paid placement. This underscores, not that search engine marketing is easy to do - but that it’s easy to spend money on.

Doing paid search well is difficult because there are so many options and you need to track them all. The optimal ad creative, landing page and price could be different for different combinations of search engine, keyword, product, geography, time (of year or of day) and for search vs content ads. You could be optimizing for immediate sales, life time value of the customer, obtaining market share or something else.

I suppose the point I’m trying to make is that paid search and search engine optimization are both difficult to do well but it’s a lot easier to spend money on paid search and that’s one of the reasons why search engine marketing spending is dominated by paid search. In my opinion the percentage spent on search engine optimization should increase over time because there is at least as much value there as in paid search. One of the challenges for those in the industry is to make it easier to spend money on SEO.

The good thing about SEO is that none of the money an SEO company charges goes to the search engines. So there is a lot more money to be made when compared to paid search. Many companies are comfortable with spending large amounts on paid search - because it is performance based. Performance based SEO services, such as Netconcepts’ GravityStream, YourAmigo, or SLI’s Site Champion service will help ensure that more of that search engine marketing spend goes on SEO.

The end of the year saw all the search engines roll out their lists of top search terms. Here are the links: Google, Yahoo, Ask Jeeves, AOL and Lycos

These are all very interesting - but the problem with them is that they are highly filtered - to remove adult terms and navigational terms. Yahoo doesn’t want to put out a press release saying that Google is one of it’s top search terms or vice versa. So what would an unfiltered list look like? It happens that we have some data to produce one.

Although most of SLI’s business is focused on providing site search and related services, we do have a portion of our business that provides services to web search companies. Typically this is providing Related Searches. When someone does a search on one of our Related Search customers’ sites (such as search.com) the query is sent to us and we return related search terms that are incorporated into the page. This means we are getting 10’s of millions of web search queries every day. By analyzing the logs we can get a list of our own. Here it is in it’s unfiltered glory:

google
rss
holiday ideas
porn
search engines
yahoo
sex
ebay
89.com
free porn
free email
white pages
find email address
people finder
email
email extractor lite
email addresses
people search
yahoo mail
hotels
ebusiness
games
business
yahoo.com
milf
email search
email address directory
mapquest
episodes
find people
myspace.com
literotica
lyrics
myspace
mail
free bulk email
dictionary
yahoo email
email directory
email marketing

So there you have it - people search for porn and uninteresting things like search engines and email providers.

I remember when we powered the search for snap.com (the old snap - not the new one) we wondered what we should possibly show people when they search for snap. Of course we let the learning decide that and the most popular result was snap.com. So these people would start off at snap.com, search for snap, and click on snap.com. You would wonder what they were thinking. That’s just the tip of the iceberg. Looking at search queries is always interesting and often perplexing. It’s even more interesting when they are queries that people are typing into your own site search. Do you know what the top search terms are for your site? You should.

Now that the holiday season is over I thought it would be interesting to look back and see what sort of holiday peak our e-commerce customers had. I took 10 customers and compared their search traffic in October November and December to that of September. This allowed me to easily compare the increases on sites with different levels of traffic. On average the traffic in December was 2.6 times higher than it was in September.

As expected, all of these customers had significantly more traffic in December and November than they had in September. That will explain why we heard so little from them over the last month - they were so busy.

Here is a video of John Battelle talking to Google in New York December 2005. John is the author of a book about Google/search.

He was introduced by Craig Nevill-Manning - a fellow New Zealander who heads the New York office.

This was entertaining - but a lot of it consisted of him reading parts from the book. So if you’ve read the book then it’s probably not worth an hour of your time. I do recommend his book - it’s a great overview of the whole search business.

I discovered podcasts when I got my first iPod (I’m now on my 2nd - have sold my mini and got a nano) and I’m hooked. I find it liberating to subscribe to podcasts that are of interest to me then listen to them where and when I want.

Podcasts are such a personal way to communicate when compared to the written word - like a blog. When you start listening to a podcast regularly you really feel like you know the podcaster.

Some of the podcasts I listen to regularly include;
Daily SearchCast
Marketwatch internet daily
Podcast 411
The Health Report
The Marketer’s Podcast
This Week in tech (TWIT).

I do find that when they talk about search on the TWIT podcast their knowledge about it is only superficial. They don’t really know what they’re talking about. This probably means they only have superficial knowledge of most of the other topics they’re talking about as well - but I still enjoy it.

I saw an article today about some new Google minis. I doubt I’m going to make it a habit of talking much about our competitors but so many people get blinded by the omnipotent Google brand I thought it was worth pointing out how we’re different.

1. Learning Search learns from the people who use it and keeps getting better
2. Learning Search helps drive traffic to your site through Site Champion (natural search) and Ad Champion (paid search)
3. We have related searches - to help people refine their search
4. Site specific search and navigation - I’m pretty sure the Goggle mini doesn’t have these
5. We’re a service - if you want something changed we have a user friendly interface called a person. It’s really easy.
6. With Learning Search you try before you buy - with a free month trial you’ll be getting the benefits before you start paying. WARNING: If you do trial there is a very high chance you’ll become a customer - nearly all sites that trial will buy.

I won’t go on - I don’t intend for this blog to sound like a sales pitch - I just had to get that off my chest.

Everyone knows that Google is the best search engine, right? I think we’re in need of objective measures of search relevance to help answer this question. Last year Danny Sullivan called for the search engines to report on relevancy. One thing lacking from this plea was an indication of what these relevancy measures might be. He had mentioned some options a few years ago but they were mainly surveys of peoples opinion which are cumbersome, expensive and subjective.

SLI has several measures of search relevancy that we make available to our site search customers. We make these available so our customers can see them improving as the search learns. These could easily be applied to web search.

The first of these is the average rank of the clicks on search results. The perfect search engine would have an average rank of 1 - that is everyone would click on the first result which would wholly satisfy them. A person clicking on a result doesn’t mean the result was good - but it does indicate that they thought from the title and description that it would be a good result. If a result doesn’t satisfy the user then common behavior is to return to the search results page and click on another - resulting in a higher average rank. This graph shows how the average rank can improve over time when learning from user click throughs.

Another simple relevance metric we offer is click through rate, i.e. the number of clicks on search results divided by the number of searches. This is somewhat crude because bad search results may encourage people to click on more than one result. Ideally this number should be 1 - i.e. everyone who searches clicks on one result. If it is less than one then it shows that a portion of people are not clicking on any search results.

Both of these could be refined more - but I haven’t seen any examples of people asking for such objective measures for web search.

We have some interesting data from one of our partners on the average rank of Google and Yahoo results. I’ll share that with you in a subsequent post. Any guesses from those of you that don’t know?

Hi,

This is the first entry in the unimaginatively named SLI Systems Blog. I’m intending for this blog to be a place where we can publish our thoughts on all things related to site search and other closely related topics. I’m sure there will be posts about web search, search engine optimization, paid search and conversion rates.

Hopefully you the reader who has stumbled across this will find something useful here. Feel free to add comments — we will read them.

All the best,
Shaun Ryan
CEO
SLI Systems

Enjoy your time here.